Governing Geoengineering Research: Principles and Process

Research to develop geoengineered climate solutions will require not just new technical knowledge, but also new societal capacity to set effective policy and align oversight with public values. Although technological governance that uses accountable democratic process requires effort, it often results in more effective policy as well as more legitimate outcomes.

A good case in point is the Swedish nuclear waste program and its contrast with the U.S. nuclear waste program.1 In 1980, Sweden voted to end nuclear power generation and began to build a repository to dispose of nuclear waste. The program proceeded with extensive public consultation and a clear a priori statement of the technical and social requirements for an appropriate site. Today, Sweden has a repository site that is scientifically sound and supported by the local population.

In contrast, the goal of the American policy was to override controversy and justify nuclear power by showing that we could store waste. Congress chose the repository site without consulting the public. Astonishingly, the site criteria were established after the site was chosen. Today, the United States does not have a successful nuclear waste storage program.2 The lack of effective debate and collaboration between policymakers, scientists, and civil society has led to a stalemate on the issue of nuclear waste management.

We cannot afford such inefficacy and deadlock in the context of geoengineering. There is too much at stake. Drawing upon experiences like these, as well as a growing body of work on geoengineering governance, we present a set of values and policies that should be signposts for geoengineering oversight.

The Precautionary Principle

The precautionary principle presents a logical dilemma in the geoengineering context: an intervention could be an important precautionary measure in the face of increasing climate disruption, yet pose serious damage to stressed ecological and political systems. But this dilemma does not lessen the importance of precaution. A precautionary stance demands, for instance, that there be a moratorium on the deployment of high-risk geoengineering technologies while research proceeds. It also counsels that we must factor uncertainty, ignorance, and irreversibility into the assessment of geoengineering experiments and applications.

International Cooperation in Science and Governance

Computer simulations indicate that China is likely to face severe water shortages that might be ameliorated by climate intervention, but the same intervention could interrupt the Indian monsoons and devastate food production. Any indication that a nation is contemplating geoengineering solely to protect national interests, especially at the expense of other countries, would and should be met with suspicion and hostility. National benefit may be the driving force for funding geoengineering research, but research programs should explicitly focus only on technology that will have significant international benefits.

Placing geoengineering research programs within defense programs would seriously undermine the spirit of cooperation needed to address climate intervention peacefully and effectively.3 On the other hand, the inclusion of international scientists in a national research program or the establishment of international research programs would be investments in the capacity to collaborate. Good cooperative relationships in geoengineering research and research governance may help to develop common norms of behavior that set the stage for collaborative decisions in the future.

Public Ownership and Open Science

Commercial interests should not be allowed to influence government regulation or deployment of geoengineering technologies where the risks of unintended harm are great and political stakes are high. High-risk geoengineering should be treated as a public endeavor because we require effective and accountable control of these technologies, untainted by commercial interests.

Governmental funders should also require open science and employ a presumption of public-sector ownership. Research contracts must obligate researchers to share certain kinds of data and to publish in accessible journals as soon as possible.

On the other hand, for technologies that are low-risk––such as technology that separates CO2 from the atmosphere—the role of the private sector might be welcome, given limitations on public resources and the advantages of competition for optimizing design. In these cases, regulation might be adequate to protect the public good and private ownership might be necessary to advance it.

Democratic Due Process

Geoengineering research is not the first case of science requiring government oversight within democratic societies. Nanotechnology, nuclear technology, and recombinant DNA all pose hazards to society. Government agencies must have ultimate authority to govern these technologies, and oversight must evolve through a publicly accountable process. Of course, researchers can initiate the process themselves. In 1975, the Nobelist Paul Berg organized a conference at Asilomar in California to discuss the potential hazards of recombinant DNA (rDNA) research and establish self-governing principles for safe science. Conferees developed a tiered structure for levels of hazard that was presented to the Department of Health and Human Services (DHHS). Based on these principles, DHHS proposed a review process for individual proposals. These rules were subject to public hearings and a review and comment period, after which they were adopted. The vibrant rDNA research program in this country is, in part, a testament to the success of this process. The organizers of a meeting on the governance of geoengineering (held at Asilomar in March 2010) consciously invoked the original rDNA meeting. Although this was a productive step toward developing governance principles for geoengineering, participants also understood the need to extend debate into civil society.4

Governments should help the public understand the ramifications of geoengineering and engage in deliberations about the research. Denmark has developed mechanisms for doing this in an institution called the Danish Board of Technology.5 The board has a menu of methods for assessing technology and deliberating with citizens, including interdisciplinary working groups; seminars; citizens’ summits, juries, and hearings; future panels; parliamentary hearings; and conferences where laypeople, experts, and politicians jointly deliberate conflicts and priorities and then vote on action plans. Policymakers are required to take the outcomes of these deliberations into account and are often required to implement the recommendations.

Independent Review

In the oversight of geoengineering experimentation, credible and independent review of research protocols will help ensure the safety, quality, and ethical integrity of the research. Review committees should include a broad array of expertise across the sciences and social sciences, and be in close communication with decision makers. Not all geoengineering research, however, should trigger the same level of review and analysis. For instance, computer modeling studies that simulate proposed interventions are almost certainly benign and should commence quickly. On the other hand, a proposal for full- or even subscale deployment with nontrivial effects would clearly require a very high level of scrutiny.

There should be an initial assessment in which experiments are targeted for either de minimis review or full review, using criteria such as degree of perturbation, degree of irreversibility, duration, and impact. The threshold for full review would vary by technology. For example, aerosol injection into the stratosphere raises completely different questions and concerns than putting small air bubbles on the surface of the ocean does. Thresholds might be determined by blue-ribbon committees convened by the National Research Council in the United States or the Royal Society in the United Kingdom, for example, and ratified by accountable regulatory authorities with public input.

Adaptive Management

Adaptive management, also known as “learning by doing,” involves learning what works and what doesn’t from both scientific and social perspectives, and then modifying research choices to reflect this new knowledge. Adaptive management is critical for geoengineering because climate intervention will likely be highly unpredictable as will the political context in which it takes place. Given these uncertainties, any climate intervention will need to adapt as new information becomes available and conditions change.

Geoengineering research should be couched in an adaptive framework from the beginning. In this way, we can gain experience in the adaptive management and at the same time learn how to do climate intervention. The exemplar Swedish nuclear waste research program established a structure a priori to accommodate modifications. First, scientific objectives were accepted by management teams and various experiments to address these objectives were reviewed and sanctioned. Then researchers predicted the outcome of the experiments. At specified time intervals, the scientific teams made formal comparisons between the predictions and their actual observations. Based on what they learned, the scientists were able to make new decisions about the next phase of an experiment and adjust their predictions for that phase. Over time, the researchers could predict the outcomes with more confidence and the waste program management learned how to adapt the program to new information, lending legitimacy to the process and making public approval more likely.

In the field of geoengineering, the iterative process may include modification of the calculus for evaluating risks and benefits, which could in turn modify experimental protocols or even terminate experiments. For this to work, it is crucial that the experiments be monitored by independent, accountable institutions not invested in the results. This independent monitoring can also provide an important vehicle for transparency and credibility.


Good governance of geoengineering research will build the technical capacity for acceptable options, coupled with the societal capacity to make good decisions about deploying them. If we succeed, these social skills may spill over into approaching other difficult climate problems. In the end, we may ask whether we are building the capacity to do geoengineering or using geoengineering research to build the capacity to solve any climate problem. If we are lucky, the answer will be the latter.